A Five Year Review OF API20E Bacteria Identification System’s Performance at a Teaching Hospital
AbstractObjectives: To assess the performance of the API20E bacteria identification system at a teaching hospital in Kenya.
Design: Retrospective study.
Setting: The microbiology laboratoryoratory of the Aga Khan University teaching Hospital.
Subjects: One thousand six hundred and fifty eight API20E records.
Main outcome measures: The accuracy in identifying the bacteria species.
Results: One thousand four hundred and forty two (87.6%) isolates had the exact identity, 199 (12%) nearest identity, and seven (0.4%) no identity. The performance varied among the species; Acinetobacter baumanii had 140 (99.3%) isolates with the exact identity and only one (0.7%) with the nearest identity compared with Aeromonas hydrophila which had five (17.2%) with exact and 24 (82.8%) with nearest.
Conclusions: The API20E system is a robust bacteria identification method which can serve small and medium clinical microbiology laboratoryoratories that may not afford automated systems. Adhering to the manufacturer’s instructions and good laboratoryoratory practice can improve the performance of this method.